Evaluating cancer therapy efficacy by studying symptomatic improvement in patient condition is a popular practice due to the lack of rapid, reliable and robust quantitative evaluation protocols for routine clinical measurement of cancer therapy efficacy. Reducing time required for arriving at decisions and reducing cost without compromising accuracy could provide potentially improved treatments especially for patients with advanced leukemia of various types. This proposal focuses on rapid quantitative assessment of cancer treatment efficacy for leukemia. The proposed approach is based on quantifying apoptosis (programmed cell death) and has been shown to be one of the most reliable and one of the earliest indicators for assessing anticancer efficacy of a drug. Several other techniques are available to detect apoptosis (Agarose gel electrophoresis, Caspase-3, TUNEL assay, Morphological estimation, Annexin V assays etc). However, these techniques are inaccurate, expensive or time consuming. A recently developed a two stage DNA diffusion assay holds promise to be an accurate, relatively inexpensive and rapid methodology for quantitative apoptosis measurement via the apoptotic index. The two stages are:(1) slide preparation and microscopy imaging, which requires approximately 1.5 hrs and (2) image analysis, which is tedious, and takes approximately 2 hours of additional intensive manual labor leading to errors in categorization of the cells. We propose to fully automate the second stage with the goal to minimize manual effort and there by reduce the human errors. This would also make the procedure more reproducible. Phase I work will involve: (1) segmentation and classification algorithm development for automatic apoptotic index calculation, and (2) characterizing the algorithm performance by inducing apoptosis in leukemia cells in culture using known apoptosis inducing agents.